Upload
duongdiep
View
219
Download
0
Embed Size (px)
Citation preview
1
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 1
www.gexcon.com © GexCon AS
Full-scale testing of dust explosions in a roller mill
Dr. Kees van Wingerden
GexCon ASBergen, Norway
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 2
www.gexcon.com © GexCon AS
Contents
Background
Description of experimental set-up
Experimental program
Design of experimental conditions usingCFD-tool DESC
Experimental results
Conclusions
2
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 3
www.gexcon.com © GexCon AS
Background
Biomass is becoming a popular fuel for power generation applications replacing coal
Grinding of biomass in a roller mill represents a hazard due to the presence of both flammable dust and possible ignition sources due to mechanical sparks, hot surfaces and smouldering combustion
Overall project aim: determination of potentialdamage to roller mill under realistic operating conditions
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 4
www.gexcon.com © GexCon AS
Contents
Background
Description of experimental set-up
Experimental program
Design of experimental conditions usingCFD-tool DESC
Experimental results
Conclusions
3
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 5
www.gexcon.com © GexCon AS
Experimental set-upGexCon test site at Sotra near Bergen, Norway
Experiments were performed in "Loesche LM 18" roller mill from a Danish power plant. The grinding capacity of the mill was 36 tons coal/hr.
Basic parts of the mill are the mill room, the classifier area and “a top” with coal feeder pipe and coal dust pipes, total volume: 23 m3.
Mill is connected to a coal feeder of 5.1 m3 via a 3.35 m long, 0.59 m diameter pipe
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 6
www.gexcon.com © GexCon AS
Roller mill
4
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 7
www.gexcon.com © GexCon AS
Experimental set-up Between mill table in bottom of the mill and mill
housing open clearance (throat) with total cross section of approximately 0.42 m2.
Three coal dust outlets of 550 mm diameter with 2 m pipe and diaphragm to simulate long pipes (> 60 m) to burners
On top of coal feeder an opening (0.27 m2) was installed to simulate the connection to a coal silo.
Primary air ducts were fixed to the mill foundation underneath the mill table (cross section of 0.85 m x 0.85 m, length 6 m)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 8
www.gexcon.com © GexCon AS
Roller mill
5
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 9
www.gexcon.com © GexCon AS
Dust dispersion Parts of suppression system (20 l bottles, opening fast acting valve by
torque motor).
Delay time choice based on DESC simulations: 900 ms
20 bar driving pressure; 1 – 3 kg dust per bottle
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 10
www.gexcon.com © GexCon AS
Diagnostics/ignition source
10 pressure transducers regularly divided over mill, primary air ducts, outlet ducts and coal feeder.
Video
Ignition source: 5 kJ Sobbe igniter positioned 0.65 m above the centre of the milling table
6
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 11
www.gexcon.com © GexCon AS
Contents
Background
Description of experimental set-up
Experimental program
Design of experimental conditions usingCFD-tool DESC
Experimental results
Conclusions
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 12
www.gexcon.com © GexCon AS
Experimental program
Effect of dust cloud location/size
Effect of dust reactivity
Effect of dust type
Effect of venting parameters
7
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 13
www.gexcon.com © GexCon AS
Dust properties
118-145 KSt-value (bar.m/s) *
7.0-7.8 Maximum explosion
pressure (bar)
2.1-2.3 Moisture content
(% w/w)
55-90 Median particle size
diameter (mm)
ValueProperty
Coal dust
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 14
www.gexcon.com © GexCon AS
Dust properties
31-5588-92112-12277KSt-value *(bar.m/s)
6.5-7.27.3-7.67.6-7-77.3Maximum explosion
pressure (bar)
6.3-6.75.5-5.85.6-6.05.4Moisture content (% w/w)
160-25095-10560-90110Median particle size
diameter (mm)
Property
Wood dust from
second source (class
III)
Wood dust from
second source (class
II)
Wood dust from
second source (class I)
Wood dust from first
source
Wood dust sample
8
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 15
www.gexcon.com © GexCon AS
Dust properties
0
20
40
60
80
100
120
140
0 500 1000 1500
Dust concentration (g/m3)
KS
t-v
alu
e (b
ar.
m/s
)Coal dust
Wood dust(source 2,class I)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 16
www.gexcon.com © GexCon AS
Contents
Background
Description of experimental set-up
Experimental program
Design of experimental conditions usingCFD-tool DESC
Experimental results
Conclusions
9
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 17
www.gexcon.com © GexCon AS
DESC simulations: geometry
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 18
www.gexcon.com © GexCon AS
Aim: obtaining turbulence conditions as when mill is used under maximum load conditions
Total air flow rate of 70,000 m3/hr through the mill
Production of 36 tons/hr of fine coal dust
10
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 19
www.gexcon.com © GexCon AS
Simulation of maximum load process conditions (flow only)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 20
www.gexcon.com © GexCon AS
Simulation of flow conditions in mill after dust injection from bottles (delay time 900 ms)
11
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 21
www.gexcon.com © GexCon AS
Contents
Background
Description of experimental set-up
Experimental program
Design of experimental conditions usingCFD-tool DESC
Experimental results
Conclusions
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 22
www.gexcon.com © GexCon AS
Experimental results: general observations
Experiments at maximum load conditions concentration wise (~500 g/m3) were avoided due to (minor) damage to mill at lower (less reactive) concentrations
Big differences between coal dust and wood dust most likely due to different dispersion properties
Big variations in wood dust samples (KSt-value determined for dust used in every single test)
12
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 23
www.gexcon.com © GexCon AS
Effect of dust cloud location: only in milling part (coal dust)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 24
www.gexcon.com © GexCon AS
Effect of dust cloud location: in entire mill and in coal feeder (coal dust)
13
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 25
www.gexcon.com © GexCon AS
Effect of dust cloud location (coal dust), measurements in mill
0.3 0.6 0.9-0.10
0.25
0.50
0.75
1.00
1.25
1.50
Time (s)
Pre
ssu
re (
bar
g)
Test no. 8; P4Test no. 9; P4Test no. 10; P4
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 26
www.gexcon.com © GexCon AS
Effect of dust cloud location (coal dust), measurements in coal feeder
0.3 0.6 0.9-0.1
0.2
0.4
0.6
0.8
1.1
Time (s)
Pre
ssu
re (
bar
g)
Test no. 8; P10Test no. 9; P10Test no. 10; P10
14
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 27
www.gexcon.com © GexCon AS
Effect of dust reactivity (dust cloud in mill only) (Wood dust KSt = 31-55 bar.m/s)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 28
www.gexcon.com © GexCon AS
Effect of dust reactivity (dust cloud in mill only)(Wood dust KSt = 88-92 bar.m/s)
15
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 29
www.gexcon.com © GexCon AS
Effect of dust reactivity (wood dust ~250 g/m3, cloud filling mill only)
0
0.5
1
1.5
2
0 20 40 60 80 100 120 140
KSt-value (bar.m/s)
Max
imu
m o
verp
ress
ure
(b
ar)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 30
www.gexcon.com © GexCon AS
Effect of dust type (test 9 coal; test 11 wood; K-value at 250 g/m3: 61 and 38 bar.m/s resp.)
0.0 0.2 0.4 0.6 0.9-0.10
0.25
0.50
0.75
1.00
1.25
1.70
Time (s)
Pre
ssu
re (
bar
g)
Test no. 9; P4Test no. 11; P4
16
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 31
www.gexcon.com © GexCon AS
Coal dust explosion (dust in entire geometry; concentration ~ 250 g/m3)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.01.0-0.10
0.25
0.50
0.75
1.00
1.25
1.70
Time (s)
Pre
ssu
re (
bar
g)
Test no. 23; P4Test no. 23; P9
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 32
www.gexcon.com © GexCon AS
Wood dust explosion (dust in entire geometry; concentration ~ 250 g/m3)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.01.0-0.1
0.5
1.0
1.5
1.9
Time (s)
Pre
ssu
re (
bar
g)
Test no. 24; P4Test no. 24; P9
17
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 33
www.gexcon.com © GexCon AS
Effect of vent opening size in coal feeder on pressure in mill; dust cloud in mill only; wood dust (KSt-value 88-92 bar.m/s)
00.20.40.60.8
11.21.41.61.8
0 0.05 0.1 0.15 0.2 0.25 0.3
Vent opening coal feeder (m2)
Ov
erp
res
su
re (
ba
r)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 34
www.gexcon.com © GexCon AS
Pressure distribution in mill for weak and strong explosions (coal dust in mill only vs coal dust in
mill and coal feeder)
Coal Mill Tests - Test 9
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10
Measurement ID
Pe
ak o
ver
pre
ssu
re (
ba
rg)
Test 9
Coal Mill Tests - Test 10
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10
Measurement ID
Pe
ak o
verp
ress
ure
(b
arg
)
Test 10
18
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 35
www.gexcon.com © GexCon AS
Contents
Background
Description of experimental set-up
Experimental program
Design of experimental conditions usingCFD-tool DESC
Experimental results
Conclusions
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 36
www.gexcon.com © GexCon AS
Conclusions
Pressures generated by wood dust explosions (KSt-value ~120 bar.m/s) occurring in a roller mill operated under conditions approaching maximum load conditions exceed 1.9 bar
Similar for coal dust explosions (KSt-value ~130 bar.m/s) exceed 1.65 bar
Important factors affecting the maximum pressure load include venting possibilities, cloud size, dust reactivity and dust dispersion properties
19
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 37
www.gexcon.com © GexCon AS
Explosion wood dust (KSt-value ~120 bar.m/s)
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 38
www.gexcon.com © GexCon AS
Acknowledgement
The financial support of this research by Vattenfall A/S and DONG Energy AS is greatly acknowledged.
The assistance of Matthijs van Wingerden, Cathrine Andersen and Knut Sømoe during the experiments is gratefully noted.
20
46th UKELG One Day Discussion Meeting “Causes, Severity and Mitigation of Aerosol and Particulate Explosions”, 22 September 2010 39
www.gexcon.com © GexCon AS
Thank you very much for yourattention!!